Clostridium difficile infection (CDI) is a leading cause of morbidity and mortality in hospitals around the world. In the US, it accounts for over $3.2 billion/year in health care costs in hospitals. Patients acquire C. difficile shortly before symptoms appear and these organisms persist in the host if antibiotic treatment had depleted the normal microbiota, creating conditions favorable for C. difficile proliferation, toxin production and disease. At least 20% of patients successfully treated for CDI symptoms have a recurrence within the next 30 days. A critical early step in many bacterial infections is attachment to specific host tissues, which guards against host defenses that may remove or kill the pathogen before it can cause disease. It is unknown how C. difficile establishes itself in the gut or evades host defenses. Importantly, it is also not known whether C. difficile makes a biofilm, a ubiquitous bacterial defense mechanism. A biofilm could not only facilitate attachment to the mucosa but also exclude critical host-defense molecules such as antimicrobial peptides and antibodies, and avoid recognition by the immune system. A C. difficile biofilm could also promote persistence and relapse (a recurrence with the same strain) if it serves as a reservoir for spores, a highly robust, dormant cell type that is resistant to antimicrobial drugs. Understanding how C. difficile interacts with host tissues, and evades host defenses and therapeutics, would greatly facilitate development of novel treatments. Our central hypothesis is that C. difficile forms a biofilm in the host gastrointestinal (GI) tract and, further, that biofilm formation facilitates infection, and protects against host defenses and antibiotic treatment. As a first step in addressing this hypothesis, we identified conditions for C. difficile biofilm formation in laboratory culture and showed that after six days, the biofilm harbors macrocolonies containing both growing and dying vegetative cells, as well as high concentrations of dormant spores. Using laser-scanning confocal microscopy, we found that polysaccharides and nucleic acids are present in the matrix. By western blot analysis, we found toxin present in the matrix, and by mass spectrometry, we identified 9 abundant matrix proteins, all of which are metabolic enzymes. Further, we used fluorescence in-situ hybridization to show that C. difficile forms communities at the mucosa of the cecum and colon in mice with CDI, indicating that C. difficile forms a biofilm in the host. C. difficile biofilms could contribute to CDI by facilitating attachment of C. difficile to appropriate locations in the colon; by resisting host defenses and antimicrobial drugs; by accumulating toxin and directing it to host tissues; and by harboring a depot of dormant spores that could facilitate relapsing disease. Here, we propose two Aims. In Aim 1, we will localize and characterize C. difficile biofilms within the GI tract in two animal models of CDI, and we will determine if the host generates an immune response to the biofilm. In Aim 2, we will examine possible mechanisms by which C. difficile biofilms may contribute to CDI.